- Kernel - mailweb.openeuler.org

[openeuler:openEuler-1.0-LTS 1597/1597] include/asm-generic/bitops/non-atomic.h:106:27: warning: array subscript 'long unsigned int[0]' is partly outside array bounds of 'u32[1]' {aka 'unsigned int[1]'}
by kernel test robot 30 Apr '25

30 Apr '25

Hi Steven, FYI, the error/warning still remains. tree: https://gitee.com/openeuler/kernel.git openEuler-1.0-LTS head: a48a35466505c8a52da57596944bd0b4485e825f commit: a6a7981cbf66d4951425d33cdce6ef39206eba83 [1597/1597] Net: ethernet: Support 3snic 3s9xx network card config: arm64-allmodconfig (https://download.01.org/0day-ci/archive/20250430/202504300130.TAvuRGbz-lkp@…) compiler: aarch64-linux-gcc (GCC) 14.2.0 reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20250430/202504300130.TAvuRGbz-lkp@…) If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot <lkp(a)intel.com> | Closes: https://lore.kernel.org/oe-kbuild-all/202504300130.TAvuRGbz-lkp@intel.com/ All warnings (new ones prefixed by >>): drivers/net/ethernet/3snic/sssnic/nic/sss_nic_ethtool_stats_api.c:492:5: warning: no previous prototype for 'sss_nic_get_io_stats_size' [-Wmissing-prototypes] 492 | u32 sss_nic_get_io_stats_size(const struct sss_nic_dev *nic_dev) | ^~~~~~~~~~~~~~~~~~~~~~~~~ In file included from arch/arm64/include/asm/bitops.h:39, from include/linux/bitops.h:20, from include/linux/kernel.h:11, from drivers/net/ethernet/3snic/sssnic/nic/sss_nic_ethtool_stats_api.c:6: In function 'test_bit', inlined from 'sss_nic_add_ethtool_link_mode' at drivers/net/ethernet/3snic/sssnic/nic/sss_nic_ethtool_stats_api.c:764:7: >> include/asm-generic/bitops/non-atomic.h:106:27: warning: array subscript 'long unsigned int[0]' is partly outside array bounds of 'u32[1]' {aka 'unsigned int[1]'} [-Warray-bounds=] 106 | return 1UL & (addr[BIT_WORD(nr)] >> (nr & (BITS_PER_LONG-1))); | ~~~~^~~~~~~~~~~~~~ drivers/net/ethernet/3snic/sssnic/nic/sss_nic_ethtool_stats_api.c: In function 'sss_nic_add_ethtool_link_mode': drivers/net/ethernet/3snic/sssnic/nic/sss_nic_ethtool_stats_api.c:759:40: note: object 'hw_mode' of size 4 759 | u32 hw_mode, u32 op) | ~~~~^~~~~~~ vim +106 include/asm-generic/bitops/non-atomic.h 4117b02132d1cf Akinobu Mita 2006-03-26 98 4117b02132d1cf Akinobu Mita 2006-03-26 99 /** 4117b02132d1cf Akinobu Mita 2006-03-26 100 * test_bit - Determine whether a bit is set 4117b02132d1cf Akinobu Mita 2006-03-26 101 * @nr: bit number to test 4117b02132d1cf Akinobu Mita 2006-03-26 102 * @addr: Address to start counting from 4117b02132d1cf Akinobu Mita 2006-03-26 103 */ 4117b02132d1cf Akinobu Mita 2006-03-26 104 static inline int test_bit(int nr, const volatile unsigned long *addr) 4117b02132d1cf Akinobu Mita 2006-03-26 105 { d05be13bcc6ec6 Jiri Slaby 2007-10-18 @106 return 1UL & (addr[BIT_WORD(nr)] >> (nr & (BITS_PER_LONG-1))); 4117b02132d1cf Akinobu Mita 2006-03-26 107 } 4117b02132d1cf Akinobu Mita 2006-03-26 108 :::::: The code at line 106 was first introduced by commit :::::: d05be13bcc6ec615fb2e9556a9b85d52800669b6 define first set of BIT* macros :::::: TO: Jiri Slaby <jirislaby(a)gmail.com> :::::: CC: Linus Torvalds <torvalds(a)woody.linux-foundation.org> -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[openeuler:OLK-5.10 2880/2880] include/linux/blk_types.h: linux/kabi.h is included more than once.
by kernel test robot 30 Apr '25

30 Apr '25

tree: https://gitee.com/openeuler/kernel.git OLK-5.10 head: b7710e01aad5e90a681267c840f4815b024b98c6 commit: dde4fe567523327d09054824030fb2d22e5009c7 [2880/2880] kabi: Add kabi reservation for storage module compiler: clang version 20.1.2 (https://github.com/llvm/llvm-project 58df0ef89dd64126512e4ee27b4ac3fd8ddf6247) If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot <lkp(a)intel.com> | Closes: https://lore.kernel.org/oe-kbuild-all/202504300111.nH5Pw0yV-lkp@intel.com/ includecheck warnings: (new ones prefixed by >>) >> include/linux/blk_types.h: linux/kabi.h is included more than once. -- >> include/linux/mm.h: linux/kabi.h is included more than once. -- >> include/linux/ioport.h: linux/kabi.h is included more than once. -- >> include/linux/device.h: linux/kabi.h is included more than once. -- >> include/linux/device/class.h: linux/kabi.h is included more than once. -- >> include/linux/cred.h: linux/kabi.h is included more than once. -- >> include/linux/swap.h: linux/kabi.h is included more than once. vim +9 include/linux/blk_types.h 8 > 9 #include <linux/kabi.h> 10 #include <linux/types.h> 11 #include <linux/bvec.h> 12 #include <linux/ktime.h> > 13 #include <linux/kabi.h> 14 -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[openeuler:OLK-5.10 2881/2881] kismet: WARNING: unmet direct dependencies detected for PTP_1588_CLOCK when selected by SXE_VF
by kernel test robot 30 Apr '25

30 Apr '25

tree: https://gitee.com/openeuler/kernel.git OLK-5.10 head: b7710e01aad5e90a681267c840f4815b024b98c6 commit: 8cee206b5558245197158bd20895f95cc28d8468 [2881/2881] Ethernet: Linkdata: Supports Linkdata ethernet Controllers config: x86_64-kismet-CONFIG_PTP_1588_CLOCK-CONFIG_SXE_VF-0-0 (https://download.01.org/0day-ci/archive/20250430/202504300151.RfBZyPgj-lkp@…) reproduce: (https://download.01.org/0day-ci/archive/20250430/202504300151.RfBZyPgj-lkp@…) If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot <lkp(a)intel.com> | Closes: https://lore.kernel.org/oe-kbuild-all/202504300151.RfBZyPgj-lkp@intel.com/ kismet warnings: (new ones prefixed by >>) >> kismet: WARNING: unmet direct dependencies detected for PTP_1588_CLOCK when selected by SXE_VF WARNING: unmet direct dependencies detected for PTP_1588_CLOCK Depends on [n]: NET [=y] && POSIX_TIMERS [=n] Selected by [y]: - SXE [=y] && NETDEVICES [=y] && ETHERNET [=y] && NET_VENDOR_LINKDATA [=y] && (X86 [=y] || ARM64) && PCI [=y] - SXE_VF [=y] && NETDEVICES [=y] && ETHERNET [=y] && NET_VENDOR_LINKDATA [=y] && (X86 [=y] || ARM64) && PCI [=y] -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[openeuler:OLK-5.10 2881/2881] kismet: WARNING: unmet direct dependencies detected for PTP_1588_CLOCK when selected by SXE
by kernel test robot 30 Apr '25

30 Apr '25

tree: https://gitee.com/openeuler/kernel.git OLK-5.10 head: b7710e01aad5e90a681267c840f4815b024b98c6 commit: 8cee206b5558245197158bd20895f95cc28d8468 [2881/2881] Ethernet: Linkdata: Supports Linkdata ethernet Controllers config: x86_64-kismet-CONFIG_PTP_1588_CLOCK-CONFIG_SXE-0-0 (https://download.01.org/0day-ci/archive/20250430/202504300001.MoxWgMgI-lkp@…) reproduce: (https://download.01.org/0day-ci/archive/20250430/202504300001.MoxWgMgI-lkp@…) If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot <lkp(a)intel.com> | Closes: https://lore.kernel.org/oe-kbuild-all/202504300001.MoxWgMgI-lkp@intel.com/ kismet warnings: (new ones prefixed by >>) >> kismet: WARNING: unmet direct dependencies detected for PTP_1588_CLOCK when selected by SXE WARNING: unmet direct dependencies detected for PTP_1588_CLOCK Depends on [n]: NET [=y] && POSIX_TIMERS [=n] Selected by [y]: - SXE [=y] && NETDEVICES [=y] && ETHERNET [=y] && NET_VENDOR_LINKDATA [=y] && (X86 [=y] || ARM64) && PCI [=y] Selected by [m]: - SXE_VF [=m] && NETDEVICES [=y] && ETHERNET [=y] && NET_VENDOR_LINKDATA [=y] && (X86 [=y] || ARM64) && PCI [=y] -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[PATCH OLK-6.6 0/7] fs/resctrl: Create l2 cache monitors
by Zeng Heng 29 Apr '25

29 Apr '25

James Morse (2): arm64: mpam: Add cpu_pm notifier to restore MPAM sysregs arm64: mpam: Restore the expected MPAM sysregs on cpuhp Zeng Heng (5): x86/resctrl: Add a handling path of default label in get_arch_mbm_state() fs/resctrl: Create l2 cache monitors fs/resctrl: Add l2 mount option to enable L2 msc arm64/mpam: Refuse cpu offline when L2 msc is enabled arm64/mpam: Refuse to enter powerdown state after L2 msc updated arch/arm64/kernel/cpufeature.c | 6 ++ arch/arm64/kernel/mpam.c | 41 ++++++++++++++ arch/x86/include/asm/resctrl.h | 2 + arch/x86/kernel/cpu/resctrl/monitor.c | 2 + drivers/platform/mpam/mpam_devices.c | 7 ++- drivers/platform/mpam/mpam_internal.h | 2 + drivers/platform/mpam/mpam_resctrl.c | 79 +++++++++++++++++++++++++-- fs/resctrl/ctrlmondata.c | 6 ++ fs/resctrl/internal.h | 2 +- fs/resctrl/monitor.c | 14 +++++ fs/resctrl/rdtgroup.c | 33 ++++++++++- include/linux/arm_mpam.h | 2 + include/linux/resctrl.h | 5 ++ include/linux/resctrl_types.h | 10 ++++ 14 files changed, 200 insertions(+), 11 deletions(-) -- 2.25.1

2 8

[PATCH OLK-6.6] ovl: don't allow datadir only
by Wang Zhaolong 29 Apr '25

29 Apr '25

From: Miklos Szeredi <mszeredi(a)redhat.com> mainline inclusion from mainline-v6.15-rc3 commit eb3a04a8516ee9b5174379306f94279fc90424c4 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IC4U5N CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- In theory overlayfs could support upper layer directly referring to a data layer, but there's no current use case for this. Originally, when data-only layers were introduced, this wasn't allowed, only introduced by the "datadir+" feature, but without actually handling this case, resulting in an Oops. Fix by disallowing datadir without lowerdir. Reported-by: Giuseppe Scrivano <gscrivan(a)redhat.com> Fixes: 24e16e385f22 ("ovl: add support for appending lowerdirs one by one") Cc: <stable(a)vger.kernel.org> # v6.7 Reviewed-by: Amir Goldstein <amir73il(a)gmail.com> Reviewed-by: Alexander Larsson <alexl(a)redhat.com> Reviewed-by: Christian Brauner <brauner(a)kernel.org> Signed-off-by: Miklos Szeredi <mszeredi(a)redhat.com> Signed-off-by: Wang Zhaolong <wangzhaolong1(a)huawei.com> --- fs/overlayfs/super.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c index 2c056d737c27..93ee57bc82ad 100644 --- a/fs/overlayfs/super.c +++ b/fs/overlayfs/super.c @@ -1178,10 +1178,15 @@ static struct ovl_entry *ovl_get_lowerstack(struct super_block *sb, if (!ofs->config.upperdir && ctx->nr == 1) { pr_err("at least 2 lowerdir are needed while upperdir nonexistent\n"); return ERR_PTR(-EINVAL); } + if (ctx->nr == ctx->nr_data) { + pr_err("at least one non-data lowerdir is required\n"); + return ERR_PTR(-EINVAL); + } + err = -EINVAL; for (i = 0; i < ctx->nr; i++) { l = &ctx->lower[i]; err = ovl_lower_dir(l->name, &l->path, ofs, &sb->s_stack_depth); -- 2.34.3

2 1

[PATCH OLK-6.6] fs: Avoid grabbing sb->s_umount under bdev->bd_holder_lock
by Wang Zhaolong 29 Apr '25

29 Apr '25

From: Jan Kara <jack(a)suse.cz> mainline inclusion from mainline-v6.7-rc1 commit fd1464105cb37a3b50a72c1d2902e97a71950af8 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IC4XMU CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… ------------------------------- The implementation of bdev holder operations such as fs_bdev_mark_dead() and fs_bdev_sync() grab sb->s_umount semaphore under bdev->bd_holder_lock. This is problematic because it leads to disk->open_mutex -> sb->s_umount lock ordering which is counterintuitive (usually we grab higher level (e.g. filesystem) locks first and lower level (e.g. block layer) locks later) and indeed makes lockdep complain about possible locking cycles whenever we open a block device while holding sb->s_umount semaphore. Implement a function bdev_super_lock_shared() which safely transitions from holding bdev->bd_holder_lock to holding sb->s_umount on alive superblock without introducing the problematic lock dependency. We use this function fs_bdev_sync() and fs_bdev_mark_dead(). Signed-off-by: Jan Kara <jack(a)suse.cz> Link: https://lore.kernel.org/r/20231018152924.3858-1-jack@suse.cz Link: https://lore.kernel.org/r/20231017184823.1383356-1-hch@lst.de Reviewed-by: Christoph Hellwig <hch(a)lst.de> Signed-off-by: Christian Brauner <brauner(a)kernel.org> Signed-off-by: Wang Zhaolong <wangzhaolong1(a)huawei.com> --- block/bdev.c | 5 +++-- block/ioctl.c | 5 +++-- fs/super.c | 50 ++++++++++++++++++++++++++++++++------------------ 3 files changed, 38 insertions(+), 22 deletions(-) diff --git a/block/bdev.c b/block/bdev.c index 3a4357e5aac6..e95578fd35d3 100644 --- a/block/bdev.c +++ b/block/bdev.c @@ -1138,13 +1138,14 @@ EXPORT_SYMBOL(lookup_bdev); void bdev_mark_dead(struct block_device *bdev, bool surprise) { mutex_lock(&bdev->bd_holder_lock); if (bdev->bd_holder_ops && bdev->bd_holder_ops->mark_dead) bdev->bd_holder_ops->mark_dead(bdev, surprise); - else + else { + mutex_unlock(&bdev->bd_holder_lock); sync_blockdev(bdev); - mutex_unlock(&bdev->bd_holder_lock); + } invalidate_bdev(bdev); } #ifdef CONFIG_DASD_MODULE /* diff --git a/block/ioctl.c b/block/ioctl.c index 71b1a185c226..712fac087f35 100644 --- a/block/ioctl.c +++ b/block/ioctl.c @@ -371,13 +371,14 @@ static int blkdev_flushbuf(struct block_device *bdev, unsigned cmd, return -EACCES; mutex_lock(&bdev->bd_holder_lock); if (bdev->bd_holder_ops && bdev->bd_holder_ops->sync) bdev->bd_holder_ops->sync(bdev); - else + else { + mutex_unlock(&bdev->bd_holder_lock); sync_blockdev(bdev); - mutex_unlock(&bdev->bd_holder_lock); + } invalidate_bdev(bdev); return 0; } diff --git a/fs/super.c b/fs/super.c index 42c9754e3add..c3da64c443dc 100644 --- a/fs/super.c +++ b/fs/super.c @@ -1428,36 +1428,51 @@ struct super_block *sget_dev(struct fs_context *fc, dev_t dev) } EXPORT_SYMBOL(sget_dev); #ifdef CONFIG_BLOCK /* - * Lock a super block that the callers holds a reference to. + * Lock the superblock that is holder of the bdev. Returns the superblock + * pointer if we successfully locked the superblock and it is alive. Otherwise + * we return NULL and just unlock bdev->bd_holder_lock. * - * The caller needs to ensure that the super_block isn't being freed while - * calling this function, e.g. by holding a lock over the call to this function - * and the place that clears the pointer to the superblock used by this function - * before freeing the superblock. + * The function must be called with bdev->bd_holder_lock and releases it. */ -static bool super_lock_shared_active(struct super_block *sb) +static struct super_block *bdev_super_lock_shared(struct block_device *bdev) + __releases(&bdev->bd_holder_lock) { - bool born = super_lock_shared(sb); + struct super_block *sb = bdev->bd_holder; + bool born; + + lockdep_assert_held(&bdev->bd_holder_lock); + lockdep_assert_not_held(&sb->s_umount); + + /* Make sure sb doesn't go away from under us */ + spin_lock(&sb_lock); + sb->s_count++; + spin_unlock(&sb_lock); + mutex_unlock(&bdev->bd_holder_lock); + born = super_lock_shared(sb); if (!born || !sb->s_root || !(sb->s_flags & SB_ACTIVE)) { super_unlock_shared(sb); - return false; + put_super(sb); + return NULL; } - return true; + /* + * The superblock is active and we hold s_umount, we can drop our + * temporary reference now. + */ + put_super(sb); + return sb; } static void fs_bdev_mark_dead(struct block_device *bdev, bool surprise) { - struct super_block *sb = bdev->bd_holder; - - /* bd_holder_lock ensures that the sb isn't freed */ - lockdep_assert_held(&bdev->bd_holder_lock); + struct super_block *sb; - if (!super_lock_shared_active(sb)) + sb = bdev_super_lock_shared(bdev); + if (!sb) return; if (!surprise) sync_filesystem(sb); shrink_dcache_sb(sb); @@ -1468,15 +1483,14 @@ static void fs_bdev_mark_dead(struct block_device *bdev, bool surprise) super_unlock_shared(sb); } static void fs_bdev_sync(struct block_device *bdev) { - struct super_block *sb = bdev->bd_holder; - - lockdep_assert_held(&bdev->bd_holder_lock); + struct super_block *sb; - if (!super_lock_shared_active(sb)) + sb = bdev_super_lock_shared(bdev); + if (!sb) return; sync_filesystem(sb); super_unlock_shared(sb); } -- 2.34.3

2 1

[PATCH OLK-6.6] fix a couple of races in MNT_TREE_BENEATH handling by do_move_mount()
by Wang Zhaolong 29 Apr '25

29 Apr '25

From: Al Viro <viro(a)zeniv.linux.org.uk> mainline inclusion from mainline-v6.15-rc4 commit 0d039eac6e5950f9d1ecc9e410c2fd1feaeab3b6 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IC4UG1 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Normally do_lock_mount(path, _) is locking a mountpoint pinned by *path and at the time when matching unlock_mount() unlocks that location it is still pinned by the same thing. Unfortunately, for 'beneath' case it's no longer that simple - the object being locked is not the one *path points to. It's the mountpoint of path->mnt. The thing is, without sufficient locking ->mnt_parent may change under us and none of the locks are held at that point. The rules are * mount_lock stabilizes m->mnt_parent for any mount m. * namespace_sem stabilizes m->mnt_parent, provided that m is mounted. * if either of the above holds and refcount of m is positive, we are guaranteed the same for refcount of m->mnt_parent. namespace_sem nests inside inode_lock(), so do_lock_mount() has to take inode_lock() before grabbing namespace_sem. It does recheck that path->mnt is still mounted in the same place after getting namespace_sem, and it does take care to pin the dentry. It is needed, since otherwise we might end up with racing mount --move (or umount) happening while we were getting locks; in that case dentry would no longer be a mountpoint and could've been evicted on memory pressure along with its inode - not something you want when grabbing lock on that inode. However, pinning a dentry is not enough - the matching mount is also pinned only by the fact that path->mnt is mounted on top it and at that point we are not holding any locks whatsoever, so the same kind of races could end up with all references to that mount gone just as we are about to enter inode_lock(). If that happens, we are left with filesystem being shut down while we are holding a dentry reference on it; results are not pretty. What we need to do is grab both dentry and mount at the same time; that makes inode_lock() safe *and* avoids the problem with fs getting shut down under us. After taking namespace_sem we verify that path->mnt is still mounted (which stabilizes its ->mnt_parent) and check that it's still mounted at the same place. From that point on to the matching namespace_unlock() we are guaranteed that mount/dentry pair we'd grabbed are also pinned by being the mountpoint of path->mnt, so we can quietly drop both the dentry reference (as the current code does) and mnt one - it's OK to do under namespace_sem, since we are not dropping the final refs. That solves the problem on do_lock_mount() side; unlock_mount() also has one, since dentry is guaranteed to stay pinned only until the namespace_unlock(). That's easy to fix - just have inode_unlock() done earlier, while it's still pinned by mp->m_dentry. Fixes: 6ac392815628 "fs: allow to mount beneath top mount" # v6.5+ Signed-off-by: Al Viro <viro(a)zeniv.linux.org.uk> Signed-off-by: Christian Brauner <brauner(a)kernel.org> Signed-off-by: Wang Zhaolong <wangzhaolong1(a)huawei.com> --- fs/namespace.c | 69 ++++++++++++++++++++++++++------------------------ 1 file changed, 36 insertions(+), 33 deletions(-) diff --git a/fs/namespace.c b/fs/namespace.c index 190fbb026ff0..f6cb0ee8cee7 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -2426,78 +2426,81 @@ static int attach_recursive_mnt(struct mount *source_mnt, static struct mountpoint *do_lock_mount(struct path *path, bool beneath) { struct vfsmount *mnt = path->mnt; struct dentry *dentry; struct mountpoint *mp = ERR_PTR(-ENOENT); + struct path under = {}; for (;;) { - struct mount *m; + struct mount *m = real_mount(mnt); if (beneath) { - m = real_mount(mnt); + path_put(&under); read_seqlock_excl(&mount_lock); - dentry = dget(m->mnt_mountpoint); + under.mnt = mntget(&m->mnt_parent->mnt); + under.dentry = dget(m->mnt_mountpoint); read_sequnlock_excl(&mount_lock); + dentry = under.dentry; } else { dentry = path->dentry; } inode_lock(dentry->d_inode); - if (unlikely(cant_mount(dentry))) { - inode_unlock(dentry->d_inode); - goto out; - } - namespace_lock(); - if (beneath && (!is_mounted(mnt) || m->mnt_mountpoint != dentry)) { + if (unlikely(cant_mount(dentry) || !is_mounted(mnt))) + break; // not to be mounted on + + if (beneath && unlikely(m->mnt_mountpoint != dentry || + &m->mnt_parent->mnt != under.mnt)) { namespace_unlock(); inode_unlock(dentry->d_inode); - goto out; + continue; // got moved } mnt = lookup_mnt(path); - if (likely(!mnt)) + if (unlikely(mnt)) { + namespace_unlock(); + inode_unlock(dentry->d_inode); + path_put(path); + path->mnt = mnt; + path->dentry = dget(mnt->mnt_root); + continue; // got overmounted + } + mp = get_mountpoint(dentry); + if (IS_ERR(mp)) break; - - namespace_unlock(); - inode_unlock(dentry->d_inode); - if (beneath) - dput(dentry); - path_put(path); - path->mnt = mnt; - path->dentry = dget(mnt->mnt_root); - } - - mp = get_mountpoint(dentry); - if (IS_ERR(mp)) { - namespace_unlock(); - inode_unlock(dentry->d_inode); + if (beneath) { + /* + * @under duplicates the references that will stay + * at least until namespace_unlock(), so the path_put() + * below is safe (and OK to do under namespace_lock - + * we are not dropping the final references here). + */ + path_put(&under); + } + return mp; } - -out: + namespace_unlock(); + inode_unlock(dentry->d_inode); if (beneath) - dput(dentry); - + path_put(&under); return mp; } static inline struct mountpoint *lock_mount(struct path *path) { return do_lock_mount(path, false); } static void unlock_mount(struct mountpoint *where) { - struct dentry *dentry = where->m_dentry; - + inode_unlock(where->m_dentry->d_inode); read_seqlock_excl(&mount_lock); put_mountpoint(where); read_sequnlock_excl(&mount_lock); - namespace_unlock(); - inode_unlock(dentry->d_inode); } static int graft_tree(struct mount *mnt, struct mount *p, struct mountpoint *mp) { if (mnt->mnt.mnt_sb->s_flags & SB_NOUSER) -- 2.34.3

2 1

[PATCH OLK-6.6] NFS: Shut down the nfs_client only after all the superblocks
by Wang Zhaolong 29 Apr '25

29 Apr '25

From: Trond Myklebust <trond.myklebust(a)hammerspace.com> mainline inclusion from mainline-v6.15-rc1 commit 2d3e998a0bc7fe26a724f87a8ce217848040520e category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IC40FO CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- The nfs_client manages state for all the superblocks in the "cl_superblocks" list, so it must not be shut down until all of them are gone. Fixes: 7d3e26a054c8 ("NFS: Cancel all existing RPC tasks when shutdown") Reviewed-by: Benjamin Coddington <bcodding(a)redhat.com> Signed-off-by: Trond Myklebust <trond.myklebust(a)hammerspace.com> Signed-off-by: Wang Zhaolong <wangzhaolong1(a)huawei.com> --- fs/nfs/sysfs.c | 22 +++++++++++++++++++++- 1 file changed, 21 insertions(+), 1 deletion(-) diff --git a/fs/nfs/sysfs.c b/fs/nfs/sysfs.c index 7b59a40d40c0..784f7c1d003b 100644 --- a/fs/nfs/sysfs.c +++ b/fs/nfs/sysfs.c @@ -12,10 +12,11 @@ #include <linux/string.h> #include <linux/nfs_fs.h> #include <linux/rcupdate.h> #include <linux/lockd/lockd.h> +#include "internal.h" #include "nfs4_fs.h" #include "netns.h" #include "sysfs.h" static struct kset *nfs_kset; @@ -226,10 +227,29 @@ static void shutdown_client(struct rpc_clnt *clnt) { clnt->cl_shutdown = 1; rpc_cancel_tasks(clnt, -EIO, shutdown_match_client, NULL); } +/* + * Shut down the nfs_client only once all the superblocks + * have been shut down. + */ +static void shutdown_nfs_client(struct nfs_client *clp) +{ + struct nfs_server *server; + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { + if (!(server->flags & NFS_MOUNT_SHUTDOWN)) { + rcu_read_unlock(); + return; + } + } + rcu_read_unlock(); + nfs_mark_client_ready(clp, -EIO); + shutdown_client(clp->cl_rpcclient); +} + static ssize_t shutdown_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { struct nfs_server *server = container_of(kobj, struct nfs_server, kobj); @@ -257,18 +277,18 @@ shutdown_store(struct kobject *kobj, struct kobj_attribute *attr, if (server->flags & NFS_MOUNT_SHUTDOWN) goto out; server->flags |= NFS_MOUNT_SHUTDOWN; shutdown_client(server->client); - shutdown_client(server->nfs_client->cl_rpcclient); if (!IS_ERR(server->client_acl)) shutdown_client(server->client_acl); if (server->nlm_host) shutdown_client(server->nlm_host->h_rpcclnt); out: + shutdown_nfs_client(server->nfs_client); return count; } static struct kobj_attribute nfs_sysfs_attr_shutdown = __ATTR_RW(shutdown); -- 2.34.3

2 1

[PATCH OLK-6.6] ksmbd: fix session use-after-free in multichannel connection
by Wang Zhaolong 29 Apr '25

29 Apr '25

From: Namjae Jeon <linkinjeon(a)kernel.org> mainline inclusion from mainline-v6.15-rc1 commit fa4cdb8cbca7d6cb6aa13e4d8d83d1103f6345db category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IC1QSM CVE: CVE-2025-22040 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- There is a race condition between session setup and ksmbd_sessions_deregister. The session can be freed before the connection is added to channel list of session. This patch check reference count of session before freeing it. Cc: stable(a)vger.kernel.org Reported-by: Sean Heelan <seanheelan(a)gmail.com> Signed-off-by: Namjae Jeon <linkinjeon(a)kernel.org> Signed-off-by: Steve French <stfrench(a)microsoft.com> Signed-off-by: Wang Zhaolong <wangzhaolong1(a)huawei.com> --- fs/smb/server/auth.c | 4 ++-- fs/smb/server/mgmt/user_session.c | 14 ++++++++------ fs/smb/server/smb2pdu.c | 7 ++++--- 3 files changed, 14 insertions(+), 11 deletions(-) diff --git a/fs/smb/server/auth.c b/fs/smb/server/auth.c index 58380a986af5..6af666749b8f 100644 --- a/fs/smb/server/auth.c +++ b/fs/smb/server/auth.c @@ -1010,13 +1010,13 @@ static int ksmbd_get_encryption_key(struct ksmbd_work *work, __u64 ses_id, if (!sess) return -EINVAL; ses_enc_key = enc ? sess->smb3encryptionkey : sess->smb3decryptionkey; - if (enc) - ksmbd_user_session_get(sess); memcpy(key, ses_enc_key, SMB3_ENC_DEC_KEY_SIZE); + if (!enc) + ksmbd_user_session_put(sess); return 0; } static inline void smb2_sg_set_buf(struct scatterlist *sg, const void *buf, diff --git a/fs/smb/server/mgmt/user_session.c b/fs/smb/server/mgmt/user_session.c index e1aabe889c61..50239175e7b4 100644 --- a/fs/smb/server/mgmt/user_session.c +++ b/fs/smb/server/mgmt/user_session.c @@ -178,11 +178,11 @@ static void ksmbd_expire_session(struct ksmbd_conn *conn) struct ksmbd_session *sess; down_write(&sessions_table_lock); down_write(&conn->session_lock); xa_for_each(&conn->sessions, id, sess) { - if (atomic_read(&sess->refcnt) == 0 && + if (atomic_read(&sess->refcnt) <= 1 && (sess->state != SMB2_SESSION_VALID || time_after(jiffies, sess->last_active + SMB2_SESSION_TIMEOUT))) { xa_erase(&conn->sessions, sess->id); hash_del(&sess->hlist); @@ -230,11 +230,12 @@ void ksmbd_sessions_deregister(struct ksmbd_conn *conn) xa_empty(&sess->ksmbd_chann_list)) { hash_del(&sess->hlist); down_write(&conn->session_lock); xa_erase(&conn->sessions, sess->id); up_write(&conn->session_lock); - ksmbd_session_destroy(sess); + if (atomic_dec_and_test(&sess->refcnt)) + ksmbd_session_destroy(sess); } } } down_write(&conn->session_lock); @@ -249,11 +250,12 @@ void ksmbd_sessions_deregister(struct ksmbd_conn *conn) ksmbd_chann_del(conn, sess); if (xa_empty(&sess->ksmbd_chann_list)) { xa_erase(&conn->sessions, sess->id); hash_del(&sess->hlist); - ksmbd_session_destroy(sess); + if (atomic_dec_and_test(&sess->refcnt)) + ksmbd_session_destroy(sess); } } up_write(&conn->session_lock); up_write(&sessions_table_lock); } @@ -309,12 +311,12 @@ void ksmbd_user_session_put(struct ksmbd_session *sess) if (!sess) return; if (atomic_read(&sess->refcnt) <= 0) WARN_ON(1); - else - atomic_dec(&sess->refcnt); + else if (atomic_dec_and_test(&sess->refcnt)) + ksmbd_session_destroy(sess); } struct preauth_session *ksmbd_preauth_session_alloc(struct ksmbd_conn *conn, u64 sess_id) { @@ -415,11 +417,11 @@ static struct ksmbd_session *__session_create(int protocol) xa_init(&sess->tree_conns); xa_init(&sess->ksmbd_chann_list); xa_init(&sess->rpc_handle_list); sess->sequence_number = 1; rwlock_init(&sess->tree_conns_lock); - atomic_set(&sess->refcnt, 1); + atomic_set(&sess->refcnt, 2); ret = __init_smb2_session(sess); if (ret) goto error; diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index b5ce09b10e3b..2074d977d5c9 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -2227,17 +2227,18 @@ int smb2_session_logoff(struct ksmbd_work *work) rsp->hdr.Status = STATUS_NETWORK_NAME_DELETED; smb2_set_err_rsp(work); return -ENOENT; } - ksmbd_destroy_file_table(&sess->file_table); down_write(&conn->session_lock); sess->state = SMB2_SESSION_EXPIRED; up_write(&conn->session_lock); - ksmbd_free_user(sess->user); - sess->user = NULL; + if (sess->user) { + ksmbd_free_user(sess->user); + sess->user = NULL; + } ksmbd_all_conn_set_status(sess_id, KSMBD_SESS_NEED_NEGOTIATE); rsp->StructureSize = cpu_to_le16(4); err = ksmbd_iov_pin_rsp(work, rsp, sizeof(struct smb2_logoff_rsp)); if (err) { -- 2.34.3

2 1